ly923976094 commented on issue #3717: kafka consume problem URL: https://github.com/apache/incubator-pinot/issues/3717#issuecomment-458420735 > `If the upstream is putting bad data and we are not able to create segments, we will (effectively) stop consuming` > That is not necessarily true, right? If a metric(revenue) value becomes -1, we can still consume and generate segments with bad data which may not be desirable. Anyhow, this is just one such instance. We can always provide workarounds for each of these scenarios but having the ability to pause/un-pause consumption is definitely worth it. I see this similar to pause/unpause feature in Helix Controller or enable/disable a partition/instance/resource. Thank you for your answers @mcvsubbu @kishoreg ，such as kishoreg says Kafka produces bad data or Kafka in this business no longer produces data and wants to continue querying data and consuming wasted machine resources (server memory, CPU, etc.)，If the address of kafka is invalid, the method to restart the server is to continue the query. However, I think this method is not advisable, because the production environment server has to be restarted. We did not use the multi-tenant function, and the real-time offline tenants are all used together
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
